In recent years, mouse and human iPS cells have been established one after another. Yamanaka et al. induced iPS cells by introducing the Oct3/4, Sox2, Klf4 and c-Myc genes into fibroblasts derived from a mouse, and forcing the cells to express the genes [WO 2007/069666 A1; Takahashi, K. and Yamanaka, S., Cell, 126: 663-676 (2006)]. Thereafter, it was revealed that iPS cells could also be produced with 3 factors other than the c-Myc gene [Nakagawa, M. et al., Nat. Biotechnol., 26: 101-106 (2008)]. Furthermore, Yamanaka et al. succeeded in establishing iPS cells by introducing the same 4 genes as those used in the mouse into human dermal fibroblasts [WO 2007/069666 A1; Takahashi, K. et al., Cell, 131: 861-872 (2007)]. On the other hand, a group of Thomson et al. produced human iPS cells using Nanog and Lin28 in place of Klf4 and c-Myc [WO 2008/118820 A2; Yu, J. et al., Science, 318: 1917-1920 (2007)].
However, the efficiency of iPS cell establishment is low at less than 1%. Especially, a problem of extremely low efficiency of iPS cell establishment occurs when they are produced by introducing 3 factors (Oct3/4, Sox2 and Klf4) other than c-Myc, which is feared to cause tumorigenesis in tissues or individuals differentiated from iPS cells, into somatic cells.
Viral vectors such as retroviruses and lentiviruses offer higher transfection efficiency than non-viral vectors, and are therefore favorable in that they enable easier generation of iPS cells. However, retroviruses and lentiviruses become integrated in the chromosome, posing a problem with safety in view of the clinical application of iPS cells. For this reason, iPS cells generated using adenovirus vectors or non-viral vectors such as plasmids without vector integration in the chromosome have been reported [Stadtfeld, M. et al., Science, 322: 945-949 (2008); Okita, K. et al., Science, 322: 949-953 (2008); Yu, J. et al., Science, 324: 797-801 (2009)]. However, these vectors are lower in iPS cell establishment efficiency than retroviruses and lentiviruses. Possibly because of the requirement of persistent high expression of reprogramming factor under iPS cell selection conditions, there are some cases in which a stable expression line having a reprogramming factor incorporated in the chromosome at a certain frequency is obtained even when using a plasmid vector, which is generally recognized as being unlikely to cause the incorporation [Okita, K. et al., Science, 322: 949-953 (2008); Kaji, K. et al., Nature, 458: 771-775 (2009)].
Hence, attempts have been made to reconcile high establishment efficiency and safety by first establishing an iPS cell using a retrovirus or lentivirus, then removing the exogenous genes from the chromosome. For example, techniques comprising a combination of a lentivirus and the Cre-loxP system have been reported [Chang, C. W. et al., Stem Cells, 27: 1042-1049 (2009); Soldner, F. et al., Cell, 136: 964-977 (2009)]. In these reports, however, a complex construct is used wherein a loxP sequence is inserted in the LTR to minimize the risk of activation of an oncogene in the vicinity by an LTR sequence outside the loxP sequence that remains after Cre recombinase treatment, and wherein another promoter such as CMV or EF1α is inserted for transcribing a reprogramming factor; therefore, there is a demand for the development of a vector that can be constructed more easily. Although exogenous nucleic acid factors can be completely eliminated of using piggyBac transposon [Kaji, K. et al., Nature, 458: 771-775 (2009)], the possibility of disturbing endogenous genes cannot be ruled out because transient integration in the genome is unavoidable.
Meanwhile, in the method involving the use of an episomal vector capable of stable self-replication outside the chromosome, in addition to the above-described low iPS cell establishment efficiency, the spontaneous clearance of the vector upon discontinuation of drug selection is of low efficiency and takes a long time [Yu, J. et al., Science, 324: 797-801 (2009)]. For this reason, there is a need for a method of removing the vector in a short time with high efficiency, while improving iPS cell establishment efficiency.
Furthermore, another problem arises in finding clinical applications for human iPS cells; the cells can become contaminated with ingredients derived from other animal species such as serum and feeder cells during iPS cell establishment and maintenance culture. It is desirable, therefore, that all operations, from reprogramming factor transfer to human iPS cell establishment and maintenance culture, be performed under “Xeno-free” conditions (no heterologous ingredients contained). However, it has been traditional practice that in human iPS cells established under virus-free conditions, heterologous ingredients are used in at least one step from reprogramming factor transfer to iPS cell establishment and maintenance culture [Okita, K. et al., Science, 322: 949-953 (2008); Yu, J. et al., Science, 324: 797-801 (2009); Kaji, K. et al., Nature, 458: 771-775 (2009)]. Meanwhile, all the human iPS cells established under Xeno-free conditions have been transfected with reprogramming genes by means of a retrovirus or lentivirus, and none of them have been prepared under virus-free conditions [Rodoriguez-Piza, I. et al., Stem Cells, 28: 36-44 (2010); Ross, P. J. et al., Stem Cells Dev., 2009 Dec. 23. (Epub ahead of print)].